Hello,
I have just now installed installed peers SIP phone on my personal computer. The reason why i checked out peers is to use it's core functions and make a SIP tunnel out of it.
I have a project which comprises of two audio servers. One takes in audio from TCP connection and the other sends audio through another TCP connection. I want to be able to communicate to this project using a phone call.
I wanted to tunnel these audio data through a SIP call.
Is that possible ? If so can some one please give me a few pointers about how to configure audio input channel in peers for this?
Many thanks,
Apurv
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
it's really hard to help if you don't give more details.
If I understand correctly, what you have is:
TCP stream ??? TCP stream
==========> server A ===> server B ==========>
and what you want is:
RTP/UDP stream TCP stream ??? TCP stream RTP/UDP stream
==============> gateway ==========> server A ===> server B ==========> gateway ==============>
SIP is just a signaling protocol, it does not tranport media stream. Media streams are transmitted over RTP (realtime transport protocol). SIP is just here to establish/teardown media sessions. Actually it (SIP) contains another protocol used to describe media session (code, ip & port), this protocol is SDP Session Description Protocol.
Hi yohannmartineau,
Thanks for your reply.
Your description of the implementation is precisely what I need. Let me explain what I am trying to do and how I plan to use peers. The project is actually to implement a dialogue manager, which uses a speech recognition engine for incoming audio and a text to speech for generating a response prompt. Previously, these speech recognizer and text to speech were configured to use system resources (mic and speaker). I am trying to make this system accessible via telephone calls. So I now have two audio servers, one to take input audio for speech recognition and a server to return audio coming from TTS. For now, these are configured to send a receive audio from TCP connection.
I actually missed the peers documentation earlier, which I am going through now. From what I understand,
i) If I can configure SDP to point to my server IP and port for audio data input, it might just solve my issue. ( Please tell me if I am correct ).
ii) TCP might not be the correct protocol to use since it's not currently supported by peers. I should try changing the implementation to UDP.
Any suggestions would appreciated. Thanks again for your help and a great project.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I have just now installed installed peers SIP phone on my personal computer. The reason why i checked out peers is to use it's core functions and make a SIP tunnel out of it.
I have a project which comprises of two audio servers. One takes in audio from TCP connection and the other sends audio through another TCP connection. I want to be able to communicate to this project using a phone call.
I wanted to tunnel these audio data through a SIP call.
Is that possible ? If so can some one please give me a few pointers about how to configure audio input channel in peers for this?
Many thanks,
Apurv
hi,
it's really hard to help if you don't give more details.
If I understand correctly, what you have is:
and what you want is:
SIP is just a signaling protocol, it does not tranport media stream. Media streams are transmitted over RTP (realtime transport protocol). SIP is just here to establish/teardown media sessions. Actually it (SIP) contains another protocol used to describe media session (code, ip & port), this protocol is SDP Session Description Protocol.
Please take a look at http://peers.sourceforge.net/files/html/peers.html
Hi yohannmartineau,
Thanks for your reply.
Your description of the implementation is precisely what I need. Let me explain what I am trying to do and how I plan to use peers. The project is actually to implement a dialogue manager, which uses a speech recognition engine for incoming audio and a text to speech for generating a response prompt. Previously, these speech recognizer and text to speech were configured to use system resources (mic and speaker). I am trying to make this system accessible via telephone calls. So I now have two audio servers, one to take input audio for speech recognition and a server to return audio coming from TTS. For now, these are configured to send a receive audio from TCP connection.
I actually missed the peers documentation earlier, which I am going through now. From what I understand,
i) If I can configure SDP to point to my server IP and port for audio data input, it might just solve my issue. ( Please tell me if I am correct ).
ii) TCP might not be the correct protocol to use since it's not currently supported by peers. I should try changing the implementation to UDP.
Any suggestions would appreciated. Thanks again for your help and a great project.
i) tcp is rarely used to transport rtp traffic.
ii) that's right, it would be easier if your server could receive rtp packets sent over udp and send rtp packets over udp.
in peers, take a look at sdp package, especially SDPManager.